SC IE N CE OF TH E TOTA L E N V IR O N ME N T 3 97 ( 20 0 8 ) 5 8–7 1
a v a i l a b l e a t w w w. s c i e n c e d i r e c t . c o m
w w w. e l s e v i e r. c o m / l o c a t e / s c i t o t e n v
Chlorinated hydrocarbon contaminants in feces
of river otters from the southern Pacific
coast of Canada, 1998–2004
John E. Elliott a,⁎, Daniel A. Guertin b , Jennifer M.E. Balke c
a
Environment Canada, Science & Technology Branch, Pacific Wildlife Research Centre, Delta, BC, Canada V4K 3N2
Department of Biological Sciences, Simon Fraser University, Burnaby, BC, Canada V5A 1S6
c
6080 Lacon Road, Denman Island, BC, Canada
b
AR TIC LE I N FO
ABS TR ACT
Article history:
Chlorinated hydrocarbon contaminants in coastal river otters (Lontra canadensis) were evaluated
Received 26 October 2007
by sampling feces (scats) collected on the south coast of British Columbia, Canada. A broad survey
Received in revised form
of industrialized areas of the Strait of Georgia region was conducted in 1998, and a subsequent
23 January 2008
survey of working harbours in 2004. Samples from 1998 were analyzed for polychlorinated
Accepted 23 January 2008
biphenyls (PCBs), organochlorine (OC) pesticides, and polychlorinated dioxins (PCDDs) and furans
Available online 22 April 2008
(PCDFs), while in 2004, chemistry was confined to ∑PCBs and OC pesticides. Concentrations of OC
pesticides were low in both years, with only dichlorodiphenyldichloroethylene (DDE; range: 0.01–
Keywords:
2.12 mg/kg lw) and hexachlorocyclobenzene (HCB; range: 0.003–0.25 mg/kg lw) detected in all
River otter
samples. In 1998, octachlorodibenzo-p-dioxin (OCDD) and other higher chlorinated PCDD/Fs were
Lontra canadensis
found in most samples, with OCDD ranging from 120 ng/kg lw in Clayoquot Sound to 19,100 ng/kg
PCBs
lw in a pooled sample from two latrines in Nanaimo. PCBs were present in all samples. In 1998
Organochlorines
geometric mean concentrations of the sum of 59 PCB congeners ranged from 0.49 mg/kg lw in
Georgia Basin
Nanaimo to 12.3 mg/kg lw in Victoria Harbour. Six years later, mean ∑PCBs remained elevated
Scat
(geometric mean 9.5 mg/kg lw) in Victoria Harbour. Geometric mean concentrations of ∑PCBs
Noninvasive sampling
from Victoria Harbour in 1998 and 2004 were N9 mg/kg lw, a published adverse effect level
for reproduction. At some latrines in both Victoria and Esquimalt Harbours, concentrations of
TCDD-toxic equivalents exceeded 1500 ng/kg lw, a value for health effects in otters that we
derived using published information. As shown in previous studies, analysis of scats
provides an efficient and non-intrusive approach to assessing contaminant threats to otter
populations, and to documenting spatial trends in residues.
© 2008 Elsevier B.V. All rights reserved.
1.
Introduction
The Georgia Basin–Puget Sound trans-boundary region is a
highly productive, temperate zone ecosystem located in
southwest British Columbia, Canada and northwest Washington, USA. Monitoring studies have documented contamination of the region by a plethora of legacy contaminants
including organochlorine (OC) pesticides, polychlorinated
biphenyls (PCBs), dioxins (PCDDs), furans (PCDFs), polycyclic
aromatic hydrocarbons (PAHs) and heavy metals (Golder
Associates, 2003). Research has linked those contaminants
to deformities and carcinomas in fish (Malins et al., 1984), as
well as reproductive and physiological effects in birds (Elliott
et al., 1989, 1996a; Sanderson et al., 1994; Gill and Elliott, 2003)
and marine mammals (Ross et al., 2000, 2004; Simms et al.,
2000).
⁎ Corresponding author. Tel.: +1 604 940 4680; fax: +1 604 946 7022.
E-mail address: john.elliott@ec.gc.ca (J.E. Elliott).
0048-9697/$ – see front matter © 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.scitotenv.2008.01.063
SC IE N CE OF TH E TOTA L E N V I RO N ME N T 3 97 ( 20 0 8 ) 5 8–7 1
The North American river otter (Lontra canadensis) inhabits
coastal regions of the Georgia Basin–Puget Sound (Melquist et al.,
2003) and has life history characteristics that make it an intriguing candidate as an ecosystem sentinel species. As sedentary,
high trophic level piscivores (Larsen, 1984; Stenson et al., 1984;
Ben-David et al., 2005) of the marine intertidal and subtidal zones,
river otters accumulate persistent contaminants in local environments and may be sensitive to dioxin-like chemicals, based on
extrapolation from the acute sensitivity of another aquatic
mustelid, the mink (Mustela vison, Jensen et al., 1977; Bleavins et
al., 1980). Although the cryptic and elusive nature of the river otter
makes it a difficult species to study directly in the wild, employing
indirect techniques that take advantage of their predictable
behavioral traits can be an efficient means of population
monitoring. River otters communicate via scent deposition at
specific locations along the land–sea interface, known as
‘latrines’ (Bowyer et al., 1995). Coastal river otters deposit feces
(scat), anal gland secretions, and urine at latrines, thus facilitating
the transfer of marine derived nutrients to the terrestrial
ecosystems (Ben-David et al., 1998, 2005). The conspicuousness
of latrines and their habitual use by otters provide opportunities
to study populations using indirect methods. In fact, the analysis
of otter scat has proven useful to measure physiological variables
such as reproductive steroid hormones (Kalz et al., 2006), as a
source of genomic DNA for individual identification (Dallas et al.,
2003; Hung et al., 2004), to evaluate and monitor the incidence of
disease (Gaydos et al., 2007), and to describe food habits (Larsen,
1984; Stenson et al., 1984; Ben-David et al., 2005). Scat samples
have also been used effectively as a way to monitor exposure of
Eurasian otters (Lutra lutra) to chlorinated hydrocarbon contamination for a number of years (e.g. Smit et al., 1994; Van den Brink
and Jansman, 2006). These methods offer advantages over
conventional techniques in that they are logistically simple and
non-intrusive to wildlife populations.
In this study, we examined the exposure of river otters to
environmental contaminants in the Georgia Basin using nonintrusive scat sampling. Our objectives were to: (1) determine
chlorinated hydrocarbon concentrations in coastal river otter
scat from urban and industrial areas of the region; (2) relate
findings to potential sources of contamination; and (3)
compare data to published criteria on critical concentrations
in otter scat and other tissues.
2.
Methods
2.1.
Study area
In 1998, sampling areas were selected on the basis of proximity
to industrial pollutant sources: Victoria Harbour, an industrial/
urban region with known PCB contamination in the inner
harbour (City of Victoria, 2001); Esquimalt Harbour, an industrial/urban area and the home port for the Canadian Pacific
naval fleet; Nanaimo, an urban area with a concentration of
forest industries; Cowichan Bay, Crofton, and Powell River,
relatively rural areas but with concentrated forest industries
(e.g. large pulp mills at Crofton and Powell River, and concentrated wood storage and processing in Cowichan Bay); and
Clayoquot Sound, a mainly wilderness reference site on the
west coast of Vancouver Island (Fig. 1).
59
In 2004, sample collection was limited to three working
harbours: Victoria Harbour was re-sampled; Vancouver Harbour, the largest industrial/urban center and main shipping
port for western Canada; and Comox Harbour, a relatively
rural reference site with a naval port, some agricultural runoff,
and some forest product processing.
2.2.
Sample collection
Fresh scats were collected from May to August in 1998 and
2004. Potential latrine sites were identified from marine
charts. The shoreline was surveyed by small watercraft and/
or by foot, and the GPS location of each latrine was recorded.
Active sites were identified by the presence of at least ten aged
scat remains and/or the presence of fresh scats, as well as
having well-established river otter trails, scrapes, and bedding
sites. All fresh scats discovered at each latrine site were colleted by hand using a latex glove or a metal spoon, combined
in acetone/hexane cleaned 50 ml jars, and stored at −20 °C.
Where mucus deposits (anal jellies) were present, they were
collected with the scats. Samples were shipped to the National
Wildlife Research Centre (NWRC), Ottawa, ON where they
were stored at −40 °C prior to processing.
Because different sample types were collected during the
two sampling periods (scat, anal jelly, and mixtures of both), we
chose to examine the relationship between chlorinated hydrocarbon levels in the two fractions. In 2006, five otter scats with a
distinct anal jelly portion disassociated from the fecal material
were collected from Victoria Harbour latrines. We assumed that
each sample was derived from an individual animal based on its
location from other excrements in the field. The two fractions
were carefully separated in the field, resulting in a total of ten
samples. Anal jelly and fecal portions were collected in 50 mL
acetone/hexane cleaned glass jars. Samples were shipped to the
National Wildlife Research Centre (NWRC), Ottawa, ON where
they were stored at −40 °C prior to processing.
2.3.
Chemical analysis
In 1998, individual fresh scats were prepared as composite
samples by latrine site, varying in volume from approximately
20 to 200 g. In 2004, some individual scats were analyzed along
with composite samples from the same latrines; composite
volumes varied according to number of fresh scats available.
Scats were placed in chemically cleaned stainless steel trays and
stirred by hand with a cleaned stainless steel rod until appearing
homogenized. Sub-samples of approximately 12 g were removed
and placed into glass jars. Split samples from 2006 were frozen in
liquid nitrogen and then homogenized to a powder in a cryogenic
Teflon ball mill (RETSCH MM301, Newtown, PA, USA).
Concentrations of OC pesticides and PCB congeners were
measured in scat samples using a multi-residue gas chromatography (GC)/mass selective detector (MSD) technique. Subsamples of 6.5 g were mixed with anhydrous Na2SO4 and let
stand for at least three hours to dry the sample. Dehydrated
scat samples were submitted to neutral lipid extraction (1:1
DCM:hexane), gel permeation chromatography, and Florisil
column chromatography as clean-up steps prior to chemical
quantification (Norstrom et al., 1988). In a variation from
standard procedures, during extraction, the sample-Na2SO4
60
SC IE N CE OF TH E TOTA L E N V IR O N ME N T 3 97 ( 20 0 8 ) 5 8–7 1
Fig. 1 – General locations of latrine sites on the south coast of British Columbia, Canada where river otter scats were collected for
environmental contaminant concentrations (1998, 2004, and 2006).
mixture was loaded onto a glass column and soaked with 50 mL of
1:1 DCM/Hexane overnight, then eluted with another 250 mL of 1:1
DCM/Hexane. Sample extracts were injected twice onto a HP5890
gas chromatograph (Hewlett Packard, Palo Alto, CA, USA) equipped
with a HP5971 mass selective detector (operated in selected ion
monitoring mode) and a 30 m DB-5 fused silica column (0.25 mm i.
d., 0.25 μm film thickness). The first injection was used to measure
OC pesticides by comparison with 21 OC standards, while the
second injection was used to measure PCB congeners against an
Aroclor 1242:1254:1260 (1:1:1) standard mixture.
OC pesticides routinely quantified using the above method
were 1,2,4,5- and 1,2,3,4-tetrachlorobenzene (TCB), pentachlorobenzene, hexachlorobenzene (HCB), octachlorostyrene (OCS),
trans- and cis-nonachlor, trans- and cis-chlordane, oxychlordane,
heptachlor epoxide (HE), p,p'-dichlorodiphenyldichloroethane
(-DDD), p,p'-dichlorodiphenyldichloroethylene (-DDE), p,p'-dichlorodiphenyltrichloroethane (-DDT), photo-mirex, mirex, α-,
β-, and γ-hexachlorocyclohexane (-HCH) and dieldrin. Fiftynine PCB congener peaks were quantified with the above
method. A 1989 diluted herring gull egg was analyzed with
each run as a reference (Turle and Collins, 1992). Results were
initially expressed on a wet weight basis uncorrected for percent
recoveries, and later converted to a lipid basis. Minimum limits
of detection varied from 0.001 to 0.0001 μg/g.
Concentrations of 18 PCDDs, 23 PCDFs and 6 non-ortho PCBs
were measured in scat samples using a high-resolution gas
chromatography/mass spectrometry (GC/MS) procedure. The
preparatory procedures (neutral extraction, gel permeation
chromatography, alumina column clean-up, Florisil column
chromatography) have been described by Letcher et al. (1996).
Quantification was achieved using a VG AutoSpec doublefocusing high-resolution MS (Waters Corp, Milford, MA, USA)
linked to a Hewlett-Packard 5890 Series II high-resolution GC
with a 30-m DB-5 fused silica column.
Isotopically-labelled (13C12) internal standards were used for
all PCDD, PCDF and non-ortho PCB congeners measured, and
corrections for percent recovery of each individual congener
were made. Herring gull egg reference samples were used to
check analytical accuracy as described above for OC and PCB
analyses. PCB congeners are described in the text using
International Union of Pure and Applied Chemistry (IUPAC)
numbers. Minimum detection limits were assessed for each
sample and are reported in the results where relevant. Lipid and
moisture content were determined using gravimetric methods.
2.4.
Statistical analysis
Samples collected in 1998 were combined as composites by
latrine site. In 2004, some individual scats were analyzed
along with composites. In both cases we treated the latrine as
the primary sampling unit. As there was significant variation
in lipid content within samples from a site, but not between
sites, data were lipid normalized as recommended by Hebert
and Keenleyside (1995). Statistical analyses were performed
using SPSS version 15.0 (Chicago, IL, USA), or by JMP version
4.0.2 (SAS Institute, Cary, NC, USA). Chlorinated hydrocarbon
SC IE N CE OF TH E TOTA L E N V I RO N ME N T 3 97 ( 20 0 8 ) 5 8–7 1
61
data were converted to natural logarithms to approximate a
normal distribution. Geometric means were calculated by site
and comparisons among sites were performed using analysis
of variance (ANOVA). When an ANOVA was significant,
multiple comparisons were made using the Tukey HSD test.
The pattern of PCBs among sites was compared by selecting
the most prevalent congeners and determining the ratio to
∑PCBs as a percentage. Percentage values were arcsine —
transformed before testing with ANOVA as described previously. Statistical significance was set at P b 0.05.
Separated anal jelly and fecal samples collected from 2006
were analyzed individually. Tests for normality (Kolmogorov–
Smirmov) were conducted and it was determined that data
transformations were unnecessary. Linear regression was
used to examine the relationship between chlorinated hydrocarbon concentrations in the fecal and anal jelly portions of
individual scat samples. Linear regression analyses were
performed using program R version 2.6.0 (Vienna, Austria).
We used the same approach as Mason et al. (1992) and Smit
et al. (1994, 1996) to derive safe and critical concentrations for
TEQs in scat. Their values were used for the variables: assimilation efficiency, conversion factor from fresh to lipid weight, food
ration, TEQ concentration in food, excretion constant and
accumulation time period. We recognize that those values are
approximate given that TEQs are based on PCDD and PCCF as well
as PCB compounds, and potential variation exists between Lutra
lutra and Lontra canadensis, but considered they were likely
adequate to derive a tentative TEQ criteria for scat.
3.
Results
3.1.
Sample type
A linear regression was performed on individual scat and anal
jelly data to determine if there was a significant relationship
between contaminant concentrations in the two fractions (Fig. 2).
A significant relation was found between scat and anal jelly
fractions for ∑PCBs (R2 =0.679, Pb 0.05, Pearson's). The slope of the
line of best-fit was not statistically different from the ideal line
with a slope of one (regression slope= 0.676, SE= 0.269, t-test of
regression slope, PN 0.05) and the intercept was not statistically
different from zero (regression y-intercept=1.44, SE =1.06, t-test
of regression y-intercept, P N 0.05).
There was also a significant relationship for total OC
pesticide contaminant concentrations in fecal material and
anal jelly portions (R2 = 0.647, P = 0.05, Pearson's). The slope of
the line of best-fit was not statistically different from one
(regression slope = 1.21, SE = 0.517, t-test of regression slope,
P N 0.05) and the intercept was not statistically different from
zero (regression y-intercept = −0.013, SE = 0.102, t-test of regression y-intercept, P N 0.05).
3.2.
Fig. 2 – Concentrations of a) ∑PCBs and b) total OC pesticides in
fecal material versus jelly portions from individual river otter
scat samples collected on the south coast of British Columbia,
Canada, 2006. The solid line is the best-fit line of the linear
regression between variables. A theoretical 1:1 relationship
is shown as a dotted line (slope = 1.0, y-intercept = 0.0). In both
cases, the slope of linear regression is not statistically different
from one, and the y-intercept is not statistically different from
zero.
measured in samples from Comox with similar concentrations
from Victoria, but significantly lower in samples from Vancouver
(P b 0.05). DDE and HCB were the only individual OC compounds
quantified in the majority of samples. As major contributors to
the total OCs, those individual compounds followed the same
spatial pattern; for example, in 1998, geometric mean concentrations of HCB and DDE in samples from Victoria Harbour were
significantly greater than samples from Nanaimo only (P b 0.05,
Table 1). In 2004, there were no significant differences among
sites for any of the individual OC pesticide compounds.
OC pesticides
3.3.
In 1998, geometric mean total OC pesticide concentrations were
highest in samples from Victoria (0.70 mg/kg lw), which were
significantly higher than mean concentrations from Nanaimo,
Powell River, and Clayoquot Sound (P b 0.05, Table 1). In 2004,
highest mean concentrations of total OC pesticides were
PCBs
In 1998, the highest geometric mean concentrations of ∑PCBs
were measured in samples from Victoria Harbour (12.3 mg/kg lw,
Table 1). ∑PCB concentrations in scats collected from Victoria
were significantly greater than concentrations at all sites other
62
SC IE N CE OF TH E TOTA L E N V IR O N ME N T 3 97 ( 20 0 8 ) 5 8–7 1
Table 1 – Mean ± standard error (and range) of total lipids and geometric mean (and range) of selected chlorinated
hydrocarbons measured in river otter scats collected from sites on the south coast of British Columbia, Canada, 1998 and
2004
Residue levels (mg/kg, lw)
Year
Location
Percent lipid
HCB
p,p'-DDE
a
a
Total OCs
a
∑PCBs
a
1998
Victoria (n = 4)
Esquimalt (n = 4)
Nanaimo (n = 5)
Powell R. (n = 5)
Cowichan (n = 2)
Clayoquot (n = 5)
2.00 ± 0.76
2.54 ± 0.36
1.0 ± 0.18
0.79 ± 0.18
1.88 ± 0.94
1.88 ± 0.23
(0.17–3.76)
(1.56–3.17)
(0.51–1.61)
(0.37–1.28)
(0.94–2.82)
(0.45–1.78)
0.05 (0.03–0.12)
0.03ab (0.01–0.05)
0.01b (0.005–0.04)
0.03ab (0.02–0.05)
0.02ab (0.01–0.02)
0.03ab (0.02–0.04)
0.29 (0.11–2.12)
0.07ab (0.02–0.23)
0.03b (0.01–0.13)
0.05ab (0.01–0.13)
0.08ab (0.07–0.08)
0.07ab (0.04–0.21)
0.70 (0.34–3.88)
0.18ab (0.13–0.37)
0.07b (0.04–0.20)
0.13b (0.05–0.24)
0.14ab (0.13–0.16)
0.14b (0.10–0.43)
12.3 (3.63–108.4)
2.11ab (1.17–2.79)
0.49b (0.23–1.42)
0.73b (0.31–1.74)
0.79b (0.69–0.89)
0.66b (0.34–1.28)
2004
Victoria (n = 7)
Vancouver (n = 7)
Comox (n = 7)
0.91 ± 0.07 (0.66–1.14)
0.91 ± 0.08 (0.59–1.15)
0.85 ± 0.15 (0.29–1.42)
0.04x (0.01–0.08)
0.03x (0.003–0.05)
0.07x (0.03–0.25)
0.17x (0.04–0.56)
0.07x (0.006–0.17)
0.26x (0.06–0.59)
0.46x (0.13–1.72)
0.17y (0.04–0.34)
0.54x (0.17–0.82)
9.48x (2.60–67.6)
3.78xy (2.41–5.12)
1.58y (0.43–4.46)
Differences between sites were analyzed by year.
One way ANOVA, Tukey HSD, P = 0.05.
Locations in each year sharing the same letters in each category are not significantly different.
than Esquimalt (F=9.71, Pb 0.001; Table 1, Fig. 3). One composite
sample collected from a Victoria latrine had a concentration of
108 mg/kg lw. In 2004, the highest geometric mean concentrations
of ∑PCBs were again measured in samples from Victoria (9.48 mg/
kg lw). ∑PCB concentrations in Victoria scat samples were
significantly higher than Comox, but not Vancouver Harbour
(Fig. 3). Samples from two Victoria Harbour latrines had elevated
concentrations of 37.6 and 67.6 mg/kg lw.
An examination of PCB congeners (Table 2) shows that
concentrations of non-ortho-PCBs were highest in samples
collected from Victoria and Esquimalt Harbours, specifically
PCB-126 and PCB -77. Those concentrations were significantly
greater (P b 0.05) than concentrations measured in samples
collected from Crofton or Clayoquot Sound. No other significant differences were observed between sampling locations.
The pattern of 33 individual PCB congeners was expressed as
a percentage of ∑PCBs (Fig. 4). The overall patterns are very
similar among sites and years, and there were no significant
differences among sites for any of the PCB congeners in both 1998
and 2004 (F = 0.37, P N 0.05; F =0.05, P N 0.05 respectively). Samples
from Victoria Harbour appear to have lower relative amounts of
the higher chlorinated congeners, particularly CBs-153, -180, and
-170/190. In contrast, in 2004, there appears to be relatively more
of the lower chlorinated congeners, CBs-66, -87 and -101/90, in
Victoria compared to Comox Harbour samples.
3.4.
PCDDs and PCDFs
Concentrations of this group of chemicals were marked by
extreme variability within and among sites. Generally, the
Fig. 3 – Geometric mean concentrations (and range) of ∑PCBs in river otter scats (expressed on a lipid basis) collected at sites on
the south coast of British Columbia, Canada and compared to published criteria for reproductive effect in otter species (Smit et al.,
1994). Sampling locations that have different subscripted letters are significantly different (P b 0.05).
0.107a (0.05–0.880)
0.017a (0.006–0.028)
0.006a (0.004–0.010)
0.010a (0.006–0.034)
NA
0.011a (0.007–0.018)
0.010a (0.009–0.011)
0.193a (0.10–2.32)
0.037a (0.023–0.048)
0.012a (0.007–0.113)
0.017a (0.009–0.034)
NA
0.05a (0.015–0.016)
0.013a (0.009–0.018)
0.296a (0.077–3.18)
0.047a (0.027–0.073)
0.014a (0.006–0.040)
0.016a (0.006–0.054)
NA
0.021a (0.021–0.021)
0.014a (0.009–0.018)
0.550a (0.136–7.24)
0.076a (0.047–0.120)
0.022a (0.012–0.40)
0.022a (0.012–0.080)
NA
0.032a (0.032–0.032)
0.023a (0.018–0.029)
2510a (493–16700)
1270a (1090–1590)
273ab (158–410)
291.2ab (186–481)
68.4b (31.6–148)
733ab (357–1510)
198b (164–238)
116a (44–713)
52.8a (39–64)
11.0a (4.2–21)
11.1a (6.0–23.0)
4.2a (3.0–6.0)
37.5a (27.0–52.0)
9.4a (8.0–11.0)
Victoria (n = 4)
Esquimalt (n = 4)
Nanaimo (n = 3)
Powell R. (n = 5)
Crofton (n = 2)
Cowichan (n = 2)
Clayoquot (n = 5)
One way ANOVA, Tukey HSD, P = 0.05.
Locations sharing the same letters in each category are not significantly different.
NA – not analyzed.
87.3a (59.5–376)
45.5a (39–55.8)
7.0a (2.0–57.5)
43.6a (39.1–52.1)
28.0a (17.8–44.0)
45.2a (42.7–47.9)
37.6a (25.0–56.7)
CB-169
CB-126
CB-77
CB-81
Location
Non-ortho-PCBs (pg/g)
934a (355–4530)
495a (386–607)
104ab (48.5–175)
92.1ab (58.6–160)
59.0b 32.8–106)
329ab (255–425)
54.9b (48.2–62.5)
CB-157
CB-156
CB-105
CB-118
Mono-ortho-PCBs (ng/g)
PCB congener
Table 2 – Geometric mean (and range) of selected polychlorinated biphenyl (PCB) congeners (lw) in river otter scats collected from sites on the south coast of British Columbia,
Canada, 1998
SC IE N CE OF TH E TOTA L E N V I RO N ME N T 3 97 ( 20 0 8 ) 5 8–7 1
63
pattern in otter scats was: OCDDN TCDFN HxCDD N HpCDFN TCDD
(Fig. 5). Concentrations of OCDD N 15,000 pg/g lw were measured
in two samples; one from Victoria Harbour and the other from
the Nanaimo area. Highest geometric mean concentrations of
TCDD (21.5 pg/g lw) and TCDF (411 pg/g lw) were measured in
samples from Esquimalt Harbour, although differences were not
significant. Relatively high concentrations of PCDFs also were
found in otter scats, with the concentration of HpCDF at one
latrine in Victoria Harbour measuring 1140 pg/g lw.
The presence of elevated 2,3,7,8-TCDF in biota from the
Georgia Basin has been related primarily to bleached-kraft
pulp mill sources (Elliott et al., 2001). The highest individual
latrine value of 2,3,7,8-TCDF (111 pg/g lw) was measured from
a sample taken near the pulp mill at Powell River. However,
high geometric mean values were found at sites in Victoria,
Esquimalt and Cowichan Bay. This chemical was not detected
in Clayoquot Sound samples.
The ratios of specific PCDFs considered as marker compounds for chemical sources were examined in more detail.
Three compounds; 12469-PnCDF, 124678-HxCDF, and 124689HxCDF, considered indicative of chlorophenolic sources
(Hagenmaier and Brunner, 1987), one compound; 234678HxCDF, considered indicative of combustion sources, and
two compounds; 12478- and 23478-PnCDF, important in
Aroclors (Hagenmaier and Brunner, 1987; Wakimoto et al.,
1988) were assessed for their spatial patterns and relative
concentrations in otter scats.
Concentrations of the combustion marker, 234678-HxCDF,
were remarkably consistent among samples. Scat samples
from Clayoquot Sound had a geometric mean concentration
of 15.2 pg/g lw, compared to 10.5 pg/g lw in Victoria, and
12.6 pg/g lw in Esquimalt Harbours. Elevated concentrations
of the chlorophenol indicators 12468-PnCDF and 124678HxCDF were also detected at some sampling locations. For
example, in scat samples collected from Victoria Harbour, the
geometric mean concentration of 12468-PnCDF was 80.8 pg/g
lw and the geometric mean of 12468-HxCDF was 82.8 pg/g lw.
Concentrations of those congeners were also elevated in
samples from Nanaimo (e.g. 124689-HxCDF, geometric mean
98.2 pg/g lw). These chemicals were not detected in Clayoquot Sound.
The Aroclor tracer 23478-PnCDF was more uniformly
present in scat samples, with a geometric mean of 16.7 pg/g
lw measured in samples from Clayoquot Sound compared to
13.9 pg/g lw at Nanaimo. The geometric mean concentration
of 23478-PnCDF, however, was markedly higher in Victoria
Harbour (56.4 pg/g lw) samples than all other sites. The other
Aroclor tracer, 12478-PnCDF, was present at elevated concentrations in Victoria Harbour (geometric mean 51.8 pg/g lw), but
non-detectable at many latrines elsewhere, including Clayoquot Sound.
3.5.
TCDD-toxic equivalents (TEQs)
The greatest geometric mean concentrations of TCDD-toxic
equivalents were measured in the samples from Esquimalt,
followed by Victoria Harbour and Cowichan Bay, all of which
were significantly greater than Clayoquot Sound, but not other
Georgia Basin sites (F = 4.79, P b 0.01; Fig. 6). In terms of relative
contribution to total TEQs, PCDFs were major at all sites, with
64
SC IE N CE OF TH E TOTA L E N V IR O N ME N T 3 97 ( 20 0 8 ) 5 8–7 1
Fig. 4 – Pattern of selected PCB congeners in river otter scats collected from latrine sites on the south coast of British Columbia,
Canada in a) 1998 and b) 2004. Data expressed as a percent of ∑PCBs and compared among sites using ANOVA.
non-ortho PCBs or PCDDs being the other key contributors,
depending on the location.
3.6.
Comparison to published literature and toxicological
criteria
Published data on PCB concentrations in scat samples from
Europe and elsewhere are summarized in Table 3, and examined below. Our data on concentrations of ∑PCBs are compared
to published effect levels by sampling location in Fig. 3, and by
individual latrine site in Fig. 7.
To derive TEQ criteria for scats, we used the tissue values of
2000 ng/kg lw as the safe level and 5000 ng/kg lw as the critical
value as described in Smit et al. (1996, pp. 87), which are based on
the depletion of Vitamin A. Therefore, we determined corresponding criteria in scats of 600 ng/kg lw for the safe value and
1500 ng/kg lw for the critical value. TEQ concentrations at
individual latrine sites are compared to the effect levels in Fig. 8.
4.
Discussion
In 1998, coastal river otters inhabiting all sampled areas of the
Georgia Basin were exposed to a variety of chlorinated
hydrocarbon contaminants, based on residues measured in
scat samples. Samples from 1998 and 2004 collected from
Victoria Harbours had ∑PCB and TEQ concentrations which
exceeded the criteria for reproductive effects in otter.
4.1.
Sample type
We differentiated two main types of samples; scats consisting
mainly of fecal material, and mucus deposits or “anal jellies.”
Often confused with secretions from the anal gland located at
the base of the rectum, the mucus deposit referred to here is
formed in the intestine and expelled through the rectum like a
scat (Kruuk, 2006). It remains unclear whether the function of
SC IE N CE OF TH E TOTA L E N V I RO N ME N T 3 97 ( 20 0 8 ) 5 8–7 1
65
Fig. 5 – Geometric mean concentrations and range of selected PCDDs and PCDFs in river otter scat collected from latrines on the
south coast of British Columbia, Canada, 1998.
the anal jelly is a response to undigested irritants in the
digestive tract such as bones and scales and/or a mechanism of
olfactory communication. Nevertheless, the contaminant concentrations could vary considerably within the two fractions
due to differences in chemical make-up. Van den Brink and
Jansman (2006) used PCB patterns to examine relationships
between ‘spraint’ (scat) and internal fat samples from individual
otters. They reported a significant linear relationship between
scat and fat samples in cases where PCB 138 and 153 accounted
for at least 42.5% of ∑PCBs, and referred to those as ‘otter
Fig. 6 – Concentrations of TCDD Toxic Equivalents (TEQs), and contribution of key components in river otter scats collected in
1998 at latrine sites on the south coast of British Columbia, Canada. Data compared among sites using ANOVA.
66
SC IE N CE OF TH E TOTA L E N V IR O N ME N T 3 97 ( 20 0 8 ) 5 8–7 1
Table 3 – Arithmetic mean (except where noted) of ∑PCBs concentrations reported in otter scat (mg/kg, lw) from various
locations
Location (year)
West England/Wales (1984–87)
Teme
Severn
Wissey
Wales (1986)
Severn
West England/Wales (1989–91)
Severn, upper
Severn, lower
Teme, upper
Teme, lower
Clun
Lugg, upper
Lugg, lower
Arrow
East England (1989–91)
Little Ouse
Thet
Black Bourn
Wissey
Wensum
Northern England (1991–92)
Northumberland
Cumbria
Esk
Annan
Galloway
Southwest England (1989–91)
North Levels
South Levels
East Devon
South Devon
Taw
Torridge
Tamar
East Cornwall
West Cornwall
Scotland (1990–91)
Upper River Clyde
Lower River Clyde
Ayrshire rivers
Inner Clyde estuary
Outer Clyde estuary
West coast
Southern Ireland (1991)
Bantry Bay
Dunmanus Bay
R. Gergus, Co. Clare
Cork Harbour
Cork City
Upper R. Lee
R. Blackwater
East Cork
Northern Ireland (1992)
River Bann
River Foyle
Belfast/Larne Loughs
Southwest Spain (1986)
Guadalquivir
Netherlands (NA)
Northeast Slovenia (1992–93)
Germany (1992)
Schleswig-Holstein
Germany (1992)
n
Mean
Range
15
30
24
5.61
0.91
15.98
b DL – 29.4
b DL – 29.3
b DL – 32.6
19
NA
2.73 (max)
89
47
83
50
55
11
6
57
3.18
8.47
2.28
5.76
2.35
1.44
8.39
1.77
0.03–62.68
0.33–44.68
0.04–15.47
0.06–40.81
0.05–11.94
0.04–5.93
0.20–67.23
0.02–7.47
96
76
75
22
23
11.31
7.77
8.71
8.07
13.97
0.09–103.9
0.37–37.93
0.20–38.06
0.56–21.15
0.79–107.48
15
6
8
21
8
7.88
8.81
4.51
2.34
0.58
0.17–34.42
4.20–13.72
0.43–8.31
0.07–4.79
0.12–1.52
25
17
14
15
23
21
21
16
15
5.11
9.44
3.70
6.04
5.71
2.51
2.42
4.15
5.52
0.23–40.16
1.20–46.55
0.51–7.90
0.86–25.58
1.27–19.59
0.54–6.07
0.18–14.31
0.76–30.46
0.93–14.41
15
11
11
19
28
29
2.06a
19.49a
5.38a
9.66a
8.45a
3.34a
0.17–4.63
5.84–180.48
2.78–11.91
0.83–53.55
0.37–74.35
0.27–28.21
11
13
5
65
15
11
62
25
2.38
0.70
0.97
1.62
2.71
2.14
0.96
1.28
0.37–6.88
0.04–4.40
0.14–3.95
0.03–6.69
0.54–7.29
0.90–3.63
0.02–4.30
0.23–8.26
7
10
7
2.54
10.86
11.23
0.33–6.82
0.45–28.90
2.54–43.14
17
7
2
NA
3.5a
0.38
1.69 (max)
1.6–6.8
0.223–0.634
3
2.93
1.54–5.71
Source
Macdonald and Mason (1988)
Delibes et al. (1991)
Mason and Macdonald (1993b)
Mason and Macdonald (1993a)
Mason (1993b)
Mason and Macdonald (1994)
Mason et al. (1992)
O'Sullivan et al. (1993)
Mason (1993a)
Delibes et al. (1991)
Smit et al. (1994)
Gutleb (1994)
Reuther and Mason (1992)
67
SC IE N CE OF TH E TOTA L E N V I RO N ME N T 3 97 ( 20 0 8 ) 5 8–7 1
Table 3 (continued)
Location (year)
n
Mean
Range
Mecklenburg-Vorpommem
Brandenburg
Niedersachsen
Bayem
Sachsen
Niedersachsen (1991/92)
Niedersachsen (before 1991)
Hungary (NA)
Austria/Czech Rep. (1990–91)
Denmark (1990)
Denmark (1991–94)
Limfjorden
Skals
Karup
Central France (2004–05)
upper Sioule River
upper Allier River
middle Allier River
lower Allier River
South Africa (1990) (Lutra maculicollis)
38
40
14
10
28
9
5
4
25
19
5.76
8.43
23.75
4.75
3.54
35.6
2.43
5.48a
4.0a
2.23a
0.56–25.80
0.35–8.43
0.99–88.42
0.32–16.47
0.15–18.53
0.99–88.42
1.35–4.80
3.42–7.78
0.42–41.0
0.36–45.95
10
10
10
b9.0
b9.0
N9.0
NR
NR
NR
17
8
24
22
4
5.06a
7.01a
7.88a
13.58a
0.12a
Source
Mason (1987)
Gutleb and Kranz (1998)
Mason and Madsen (1993)
Elmeros and Leonards (1994)
Lemarchand et al. (2007)
NR
NR
NR
NR
0.05–0.24
Mason and Rowe-Rowe (1992)
Samples obtained from Lutra lutra unless specified otherwise.
geometric mean.
b DL – less than detection limit.
NR – not reported.
a
spraints’ that reflect the internal PCB concentrations of otters, as
opposed to ‘fish spraints’ that reflect concentrations in otter
prey. For the purposes of the analysis here, we have not
undertaken a detailed statistical analysis of PCB or other
contaminant patterns in scat and anal jellies. We have opted
to rely on our direct comparisons of chemical concentrations
between anal jelly and fecal portions of individual scat samples.
Although sample size was low, regression analyses indicates
that PCB and OC pesticide concentrations in fecal material were
quantitatively similar to those in anal jellies; therefore, we have
treated all sample types as equivalent and comparable.
4.2.
OC pesticide exposure
Although there were significant differences between sampling sites, OC pesticide concentrations measured in river
otter scats were generally low throughout the Georgia Basin in
both 1998 and 2004, and were similar to those reported at
relatively clean sites in Britain for example (Mason 1993b;
Mason and Macdonald 1994). This is consistent with other
studies measuring OC pesticide concentrations in liver
samples obtained from river otter and mink inhabiting
the lower Fraser River basin in British Columbia (Elliott et al.,
1999; Harding et al., 1999). In contrast, concentrations of
PCBs pose a greater threat to river otter populations throughout the region.
4.3.
PCB exposure and significance
Conclusions about the negative impacts of PCBs on Eurasian
otter populations have been based on the assumption that
otters are as sensitive to PCB exposure as mink, and that
50 mg/kg lw in fat tissue was a critical level above which mink
reproduction could be severely affected (Mason and Macdonald, 1994). To estimate tissue concentrations in otters, Mason
et al. (1992) applied a single compartment model, originally
developed by de Vries (1989, cited in Mason et al., 1992). Target
values to assess the significance of PCB levels in otter scat
were determined based on a back-calculation of the model,
assuming that the contaminants measured in scats are
derived entirely from unassimilated chemicals in the diet.
Model parameters were derived from experiments with mink
and adapted for the otter when possible. Target values were
further assessed by Smit et al. (1994), and they recommended
use of a no effect level (NOEL) of 4 mg/kg lw ∑PCBs, a range of
concern from 9–16 mg/kg lw, and a critical value of 16 mg/kg lw
∑PCBs in scat. This model has been used in a variety of studies
in Europe, and estimates of low otter population density were
correlated with areas where PCB levels in otter scats exceeded
the model's level of concern (Mason and Macdonald, 1993a,b,
1994). Despite potential differences in the physiology and
behaviour of Lutra lutra and Lontra canadensis, those criteria
provide a basis to assess the toxicological significance of PCB
levels in North American river otter scat. We have opted to use
the critical value of 9 mg/kg lw for potential PCB effects on
otter reproduction as suggested by de Vries (1989, cited in Smit
et al., 1994), based on the geometric mean tissue concentration
of 30 mg/kg lw in a Swedish otter population (with an
arithmetic mean of 50 mg/kg lw). A study of Swedish otter
populations in relation to chlorinated hydrocarbon exposure
appears to validate the use of that lower threshold value (Roos
et al., 2001). It should be recognized, however, that the value of
30 mg/kg lw ∑PCBs may not be particularly conservative, given
that tissue concentrations of approximately 12 mg/kg lw in
adult female mink were associated with reproductive effects
in a more recent chronic dosing study (Brunstrom et al., 2001).
68
SC IE N CE OF TH E TOTA L E N V IR O N ME N T 3 97 ( 20 0 8 ) 5 8–7 1
Fig. 7 – Concentrations of ∑PCBs in separate scat samples collected from individual river otter latrines sites on the south coast of
British Columbia, Canada in a) 1998 and b) 2004 compared to published criteria for reproductive effects in otter species.
Fig. 8 – Concentrations of TEQs in scats collected from individual river otter latrine sites on the south coast of British Columbia,
Canada (1998), compared to criteria for reproductive effects in otter species.
SC IE N CE OF TH E TOTA L E N V I RO N ME N T 3 97 ( 20 0 8 ) 5 8–7 1
A study of Swedish otter populations provided some support
for that lower threshold value (Roos et al., 2001).
∑PCB concentrations in scats from most sampled locations,
including Vancouver Harbour, were relatively low and below
the putative safe level. Comparisons with published data
should be made with caution, however, given variation in
factors such as sampling and pooling, extraction, instrumentation, and statistical analysis. Nevertheless, we have summarized the literature values for comparative purposes
(Table 3).
Scat samples collected from coastal southwestern British
Columbia, with the exception of Victoria Harbour, had ∑PCB
concentrations similar to concentrations measured in Eurasian otter scats collected from areas of low human density in
Wales and southwest Spain (Delibes et al., 1991), and from
otter population strongholds in Slovenia, Austria, Czech
Republic (Gutleb, 1994; Gutleb and Kranz, 1998), Denmark
(Mason and Madsen, 1993; Elmeros and Leonards, 1994), and
coastal southern Ireland (O'Sullivan et al., 1993).
While the majority of scat samples in both surveys had
low concentrations of ∑PCBs, geometric mean concentrations of ∑PCBs measured in samples from Victoria in 1998
(12.3 mg/kg lw) and 2004 (9.48 mg/kg lw) exceeded the level
of concern (9–16 mg/kg lw) and our suggested critical level
(9 mg/kg lw) for the health or reproductive ability of otters,
based on ∑PCBs alone (Fig. 3). In fact, out of all sites sampled,
only samples collected from latrines in Victoria Harbour in
both 1998 (n = 2 latrines) and 2004 (n = 3 latrines) exceeded this
level (Fig. 7). Those values are slightly greater than the maximum allowable concentration if combined with the total OCs
as suggested by Mason and Macdonald (1993b). The elevated
concentrations measured in scats from Victoria Harbour are
comparable to those from aquatic systems throughout Europe
that were considered too contaminated to support and sustain
viable populations of Eurasian otters (Mason et al., 1992;
Reuther and Mason, 1992; Mason, 1993a,b).
In contrast, concentrations of PCBs in livers from otters
from the sparsely inhabited Shetland Isles of Scotland were
comparable to those from more contaminated sites in coastal
British Columbia, and yet the Shetland populations were
described as thriving (Kruuk and Conroy, 1996) with high
fecundity and survival attributed to overall quality of habitat,
food availability, and the relative lack of other anthropogenic
stressors in the system. Otters inhabiting large urban
harbours such as Victoria, however, are subject to many
stressful and resource-limiting conditions associated with
degraded habitat quality, including a modified prey base,
human and pet disturbance, auto, boat, and seaplane traffic,
pollution by domestic sewage, parasites, and other chemical
contaminants such as PAHs and heavy metals (City of
Victoria, 2001). Such factors may combine to create biological
sinks for river otter populations. Further examination of the
health and population status of river otter populations of the
British Columbia coast is required to determine the cumulative effects of co-exposure to contaminants and other
stressors. In particular, further study of river otters inhabiting
the waters of Victoria Harbour is necessary to understand the
extent to which those animals are impacted by PCBs and
other contaminants. An investigation is underway using fecal
DNA to determine the number of otters potentially impacted
69
by elevated contaminant levels, and the relevant aspects of
their population dynamics (Guertin, unpublished data). The
goal is to determine whether the population is entirely or
partially resident or transient, the degree to which animals
are related, and whether there is evidence of successful local
reproduction, critical to determining whether the harbour is a
sink for otters moving in from neighboring less polluted
environments.
4.4.
PCDD, PCDF and TEQ exposure
In addition to PCB and OC pesticide contamination, river otters
were exposed to substantial concentrations of chlorinated
dioxins and furans at some locations in the Georgia Basin, at
least in 1998. The pattern of PCDDs and PCDFs observed in this
study, particularly the domination of OCDD, is similar to that
reported in river otter liver tissue obtained from trappers along
the Columbia River downstream of Portland, OR, USA (Elliott
et al., 1999). Chlorophenolic wood preservatives were considered the source of the PCDD/F contaminants in the lower
Columbia River system. In addition, the presence of 12468PnCDF, 124678-HxCDF, and 124689-HxCDF (indicative of
chlorophenolic wood preservatives), and relatively lower
amounts of 234678-HxCDF (indicative of combustion, Hagenmaier and Brunner, 1987; Wakimoto et al., 1988) in otter scats
from Victoria and Esquimalt Harbours is consistent with
chlorophenol sources, as determined previously for ospreys
(Pandion haliaetus) from the Columbia River (Elliott et al., 1998)
and beluga whales (Delphinapterus leucas) from the St. Lawrence River estuary (Muir et al., 1996). The presence of elevated
concentrations of the two Aroclor tracers (ibid.) 12478- and
23478-PnCDF in scats from Victoria Harbour is indicative of
Aroclor pollution in the harbour. This is also clearly illustrated
by the high PCB concentrations measured.
In contrast to the data from the early 1990s in resident
marine-foraging bird populations (Elliott et al., 1996b, 2001;
Harris et al., 2003), there was minimal evidence of elevated
TCDD or TCDF in scat samples collected in the vicinity of the
major bleached-kraft pulp mills at Crofton, Nanaimo, or
Powell River. A latrine site at Powell River had the highest
individual TCDF contamination, demonstrating some residual contamination. TCDF was also found at relatively high
concentrations in scats from both Esquimalt and Cowichan.
The source is likely associated with the substantial chlorophenolic contamination indicated by the other PCDD and
PCDF contaminants present at high concentrations in otter
scat. That TCDF was present in scat samples contrasts with
the previous analyses, where no TCDF was detected in the
livers of otters collected in the same general area (Elliott et al.,
1999). TCDF is readily metabolized (Van den Berg et al., 1994),
likely accounting for its absence from liver samples, while its
presence in scat is probably due to unassimilated prey
material, particularly crab or fish. TCDD, at 11 pg/g wet
weight was reported previously in one otter liver collected
close to the pulp mill at Castlegar, British Columbia.
In 1998, TEQ concentrations at some latrine sites in Victoria
and Esquimalt Harbours exceeded our calculated threshold of
1500 ng/g lw, providing further support for potential toxicological impacts on otters using the harbours. Non- and monoortho-PCBs contributed substantially to TEQs at both sites,
70
SC IE N CE OF TH E TOTA L E N V IR O N ME N T 3 97 ( 20 0 8 ) 5 8–7 1
and data from 2004 show ongoing PCB contamination, and
therefore TEQ contamination in Victoria Harbour.
5.
Summary
Scat samples collected around the Strait of Georgia region of
southwest British Columbia show that coastal river otters were
contaminated with a variety of chlorinated hydrocarbons,
particularly PCBs, PCDDs and PCDFs, and that exposure was
significantly higher than samples from Clayoquot Sound on the
west coast of Vancouver Island. Mean concentrations of ∑PCBs
in scat samples from Victoria Harbour and TEQ concentrations
in samples from individual latrines in Victoria and Esquimalt
Harbours exceeded criteria for reproductive effects developed
for the Eurasian otter. Further work is needed to determine if
exposure to those toxic and endocrine-disrupting substances is
impacting the health of river otter populations in Victoria
Harbour or elsewhere in the Georgia Basin–Puget Sound region.
Acknowledgements
This research was funded by the Georgia Basin Action Plan and
the Canadian Wildlife Service of Environment Canada, by the
British Columbia Habitat Conservation Trust Fund, and by the
American Society of Mammalogists. C. Gill, S. Lee, C. Nelson
and C. Engelstoft assisted with collection of scats. G. Savard
prepared samples for analysis. H. Won and A. Idrissi conducted
organochlorine pesticide and PCB analysis. M. Simon and
M. Mulvihill performed the PCDD, PCDF and non-ortho PCB
analyses. S. Lee assisted with graphics preparation.
REFERENCES
Ben-David M, Bowyer RT, Duffy LK, Roby DD, Schell DM. Social
behavior and ecosystem processes: river otter latrines and
nutrient dynamics of terrestrial vegetation. Ecology
1998;79:2567–71.
Ben-David M, Blundell GM, Kern JW, Maier JAK, Brown ED, Jewett
SC. Communication in river otters: creation of variable
resource sheds for terrestrial communities. Ecology
2005;86:1331–45.
Bleavins MR, Aulerich RJ, Ringer RK. Polychlorinated biphenyls
(aroclors 1016 and 1242): effects on survival and reproduction
in mink and ferrets. Arch Environ Contam toxicol
1980;9:627–35.
Bowyer TB, Testa JW, Faro JB. Habitat selection and home ranges of
river otters in a marine environment: effects of the Exxon
Valdez oil spill. J. Mammal. 1995;76:1–11.
Brunstrom B, Lund B-O, Bergman A, Asplund L, Athanassiadis I,
Athanasiadou M, et al. Reproductive toxicity in mink (Mustela
vison) chronically exposed to environmentally relevant
polychlorinated biphenyl concentrations. Environ Toxicol
Chem 2001;20:2318–27.
City of Victoria. Victoria Harbour plan; 2001. Victoria, BC, Canada,
51 pp.
Dallas JF, Coxon KE, Sykes T, Chanin PRF, Marshall F, Carss DN,
et al. Similar estimates of population genetic composition and
sex ratio derived from carcasses and faeces of Eurasian otter
(Lutra lutra). Mol Ecol 2003;12:275–82.
Delibes M, Macdonald SM, Mason CF. Seasonal marking, habitat
and organochlorine contamination in otters (Lutra lutra): a
comparison between catchments in Andalusia and Wales.
Mammalia 1991;55:567–78.
Elliott JE, Butler RW, Norstrom RJ, Whitehead PE. Environmental
contaminants and reproductive success of great blue herons
(Ardea Herodias) in British Columbia, 1986–87. Environ Pollut
1989;59:91–114.
Elliott JE, Norstrom RJ, Lorenzen A, Hart LE, Philbert H, Kennedy SW,
et al. Biological effects of polychlorinated dibenzo-p-dioxins,
dibenzofurans, and biphenyls in bald eagle (Haliaeetus
leucocephalus) chicks. Environ Toxicol Chem 1996a;15:782–93.
Elliott JE, Wilson LK, Langelier KW, Norstrom RJ. Bald eagle
mortality and chlorinated hydrocarbon contaminants in livers
from British Columbia, Canada, 1989–1994. Environ Pollut
1996b;94:9–18.
Elliott JE, Machmer MM, Henny CJ, Wilson LK, Norstrom RJ.
Contaminants in ospreys from the Pacific Northwest: I.
Trends and patterns in polychlorinated dibenzo-p-dioxins and
dibenzofurans in eggs and plasma. Arch Environ Contam
Toxicol 1998;35:620–31.
Elliott JE, Henny CJ, Harris ML, Wilson LK, Norstrom RJ. Chlorinated
hydrocarbons in livers of American mink (Mustela vison) and
river otter (Lutra canadensis) from the Columbia and Fraser River
basins, 1990–1992. Environ Monit Assess 1999;57:229–52.
Elliott JE, Harris ML, Wilson LK, Whitehead PE, Norstrom RJ.
Monitoring temporal and spatial trends in polychlorinated
dibenzo-p-dioxins (PCDDs) and dibenzofurans (PCDFs) in eggs
of great blue heron (Ardea herodias) on the coast of British
Columbia, Canada, 1983–1998. Ambio 2001;30:416–28.
Elmeros M, Leonards P. Aspect of PCB levels in Danish otters (Lutra
lutra) and notes on the levels in stone martens (Martes foina).
IUCN Otter Spec Group Bull 1994;10:34–40.
Gaydos JK, Miller WA, Gilardi KVK, Melli A, Schwantje H, Engelstoft
C, et al. Cryptosporidium and Giardia in marine-foraging river
otters (Lontra canadensis) from the Puget Sound Georgia Basin
ecosystem. J Parasitol 2007;93:198–202.
Gill CE, Elliott JE. Influence of food supply and chlorinated
hydrocarbon contaminants on breeding success of bald eagles.
Ecotoxicology 2003;12:95–111.
Golder Associates (formally EVS Consultants). Status, trends and
effects of toxic contaminants in the Puget Sound environment.
Prepared for the Puget Sound Action Team. Olympia, WA; 2003.
72 pp.
Gutleb AC. Heavy metals, organochlorpesticides and PCBs in
spraints of the otter (Lutra lutra) from northeastern Slovenia.
IUCN Otter Spec Group Bull 1994;10:31–4.
Gutleb AC, Kranz A. Estimation of polychlorinated biphenyl (PCB)
levels in livers of the otter(Lutra lutra) from concentrations in
scats and fish. Water Air Soil Pollut 1998;106:481–91.
Hagenmaier H, Brunner H. Isomerspecific analysis of
pentachlorophenol and sodium pentachlorophenate for
2,3,7,8-Substituted PCDD and PCDF at sub-ppb levels.
Chemosphere 1987;16:1759–64.
Harding LE, Harris ML, Stephen CR, Elliott JE. Reproductive and
morphological condition of wild mink (Mustela vison) and river
otters (Lutra canadensis) in relation to chlorinated hydrocarbon
contamination. Environ Health Perspect 1999;107:141–7.
Harris ML, Wilson LK, Norstrom RJ, Elliott JE. Egg concentrations of
polychlorinated dibenzo-p-dioxins and dibenzofurans in double-crested (Phalacrocorax auritus) and Pelagic (P. pelagicus)
cormorants from the Strait of Georgia, Canada, 1973-1998.
Environ Sci Technol 2003;37:822–31.
Hebert CE, Keenleyside KA. To normalize or not to normalize? Fat
is the question. Environ Toxicol Chem 1995;14:801–8.
Hung CM, Li SH, Lee LL. Faecal DNA typing to determine the
abundance and spatial organization of otters (Lutra lutra) along
two stream systems in Kinmen. Anim Conserv 2004;7:301–11.
SC IE N CE OF TH E TOTA L E N V I RO N ME N T 3 97 ( 20 0 8 ) 5 8–7 1
Jensen S, Kihlstrom JE, Olsson M, Lundberg C, Orberg J. Effects of
PCB and DDT on mink (Mustela vision) during the reproductive
season. Ambio 1977;6:239.
Kalz B, Jewgenow K, Fickel J. Structure of an otter (Lutra lutra)
population in Germany – results of DNA and hormone analyses
from faecal samples. Mamm Biol 2006;71:321–35.
Kruuk H. Otters: ecology, behaviour and conservation. New York:
Oxford University Press; 2006. 265 pp.
Kruuk H, Conroy JWH. Concentrations of some organochlorines in
otters (Lutra lutra L.) in Scotland: implications for populations.
Environ Pollut 1996;92:165–71.
Larsen DN. Feeding habits of river otters in coastal southeastern
Alaska. J Wildl Manage 1984;48:1446–52.
Lemarchand C, Amblard C, Souchon Y, Berny P. Organochlorine
compounds (pesticides and PCBs) in scats of the European otter
(Lutra lutra) from an actual expanding population in central
France. Water Air Soil Pollut 2007;186:55–62.
Letcher RJ, Norstrom RJ, Lin S, Ramsay MA, Bandiera SM.
Immunoquantitation and microsomal monooxygenase
activities of hepatic cytochromes P4501A and p4502B and
chlorinated hydrocarbon contaminant levels in polar bear
(Ursus maritimus). Toxicol Appl Pharmacol 1996;137(2):127–40.
Macdonald SM, Mason CF. Observations on an otter population in
decline. Acta Theriol 1988;33:415–34.
Malins DC, McCain BB, Brown DW, Chan SL, Myers MS, Landahl JT,
et al. Chemical pollutants in sediments and diseases of bottom
dwelling fish in Puget Sound, Washington. Environ Sci Technol
1984;18:705–13.
Mason CF. Organochlorine residues in otter spraints from
Hungary. IUCN Otter Spec Group Bull 1987;2:50–1.
Mason CF. PCB and organochlorine pesticide residues in a sample
of otter (Lutra lutra) spraints from northern Ireland. Biol
Environ-Proc R Ir Acad 1993a;93B:111–2.
Mason CF. Regional trends in PCB and pesticide contamination in
northern Britain as determined in otter (Lutra lutra) scats.
Chemosphere 1993b;26:941–4.
Mason CF, Rowe-Rowe DT. Organochlorine pesticide residues and
PCBs in otter scats from Natal. S Afr J Wildl Res 1992;22:29–31.
Mason CF, Macdonald SM. Impact of organochlorine pesticide
residues and PCBs on otters (Lutra lutra) in eastern England. Sci
Total Environ 1993a;138:147–60.
Mason CF, Macdonald SM. Impact of organochlorine pesticide
residues and PCBs on otters (Lutra lutra): a study from western
Britain. Sci Total Environ 1993b;138:127–45.
Mason CF, Madsen AB. Organochlorine pesticide residues and
PCBs in Danish otters (Lutra lutra). Sci Total Environ
1993;133:73–81.
Mason CF, Macdonald SM. PCBs and organochlorine pesticide
residues in otters (Lutra lutra) and in otter spraints from SW
England and their likely impact on populations. Sci Total
Environ 1994;144:305–12.
Mason CF, Macdonald SM, Bland HC, Ratford J. Organochlorine
pesticide and PCB contents in otter (Lutra lutra) scats from
western Scotland. Water Air Soil Pollut 1992;64:617–26.
Melquist WE, Polechla PJ, Toweill D. River otter. In: Feldhamer GA,
Thompson BC, Chapman JA, editors. Wild mammals of North
America. 2nd ed. Baltimore: Johns Hopkins University Press;
2003. p. 708–34.
Muir DCG, Ford CA, Rosenberg B, Norstrom RJ, Simon M, Beland P.
Persistent organochlorines in beluga whales (Delphinapterus
71
leucas) from the St Lawrence River estuary. 1. Concentrations
and patterns of specific PCBs, chlorinated pesticides and
polychlorinated dibenzo-p-dioxins and dibenzofurans. Environ
Pollut 1996;93:219–34.
Norstrom RJ, Simon M, Muir DCG, Schweinsburg RE. Organochlorine
contaminants in Arctic marine food chains: identification,
geographical distribution, and temporal trends in polar bears.
Environ Sci Technol 1988;22:1063–71.
O'Sullivan WM, Macdonald SM, Mason CF. Organochlorine pesticide
residues and PCBs in otter spraints from southern Ireland.
Biology and Environment. Proc R Ir Acad 1993;93B:55–7.
Reuther C, Mason CF. Erste Ergebnisse von Kotanalysen zur
Schadstoff-Belastung deutscher otter. Habitat 1992;7:7–21.
Roos A, Greyerz E, Olsson M, Sandegren F. The otter (Lutra lutra) in
Sweden — population trends in relation to ΣDDT and total PCB
concentrations during 1968–99. Environ Pollut 2001;111:457–69.
Ross PS, Ellis GM, Ikonomou MG, Barrett-Lennard LG, Addison RF.
High PCB concentrations in free-ranging Pacific killer whales,
Orcinus orca: effects of age, sex, and dietary preference. Mar
Pollut Bull 2000;40:504–15.
Ross PS, Jeffries SJ, Yunker MB, Addison RF, Ikonomou MG,
Calambokidis JC. Harbor seals (Phoca vitulina) in British Columbia,
Canada, and Washington State, USA, reveal a combination of
local and global polychlorinated biphenyl, dioxin, and furan
signals. Environ Toxicol Chem 2004;23:157–65.
Sanderson JT, Elliott JE, Norstrom RJ, Whitehead PE, Hart LE, Cheng
KM, et al. Monitoring biological effects of polychlorinated
dibenzo-p-dioxins, dibenzofurans, and biphenyls in great blue
heron chicks (Ardea herodias) in British Columbia. J Toxicol
Environ Health 1994;41:435–50.
Simms W, Jeffries S, Ikonomou MG, Ross PS. Contaminant related
disruption of vitamin A dynamics in free-ranging harbor seal
(Phoca vitulina) pups from British Columbia, Canada, and
Washington State, USA. Environ Toxicol Chem 2000;19:2844–9.
Smit MD, Leonards PEG, van Hattum AGM, de Jongh AWJJ. PCBs in
European otter (Lutra lutra) populations, report nr 4-97/7.
Amsterdam: Institute for Environmental Studies, Vrije
Universiteit; 1994. 45 pp.
Smit MD, Leonards PEG, Murk AJ, de Jongh AWJJ, van Hattum B.
Development of otter-based quality objectives for PCBs.
Amsterdam: Institute for Environmental Studies, Vrije
Universiteit; 1996. 129 pp.
Stenson GB, Badgero GA, Fisher HD. Food habits of the river otter
(Lutra canadensis) in the marine environment of British
Columbia. Can J Zool 1984;62:88–91.
Turle R, Collins B. Validation of the use of pooled samples for
monitoring of contaminants in wildlife. Chemosphere
1992;25:463–9.
Van den Berg M, Dejongh J, Poiger H, Olson JR. The toxicokinetics
and metabolism of polychlorinated dbenzo-p-dioxins (PCDDs)
and dibenzofurans (PCDFs) and their relevance for toxicity. Crit
Rev Toxicol 1994;24:1–74.
Van den Brink NW, Jansman HAH. Applicability of spraints for
monitoring organic contaminants in free ranging otters
(Lutra lutra). Environ Toxicol Chem 2006;25:2821–6.
Wakimoto T, Kannan N, Ono M, Tatsukawa R, Masuda Y. Isomer
specific determination of polychlorinated dibenzofurans in
Japanese and American polychlorinated biphenyls.
Chemosphere 1988;17:743–50.